Problems using Multi-GPT on Mac (Pro M1 and Mini)

[zepef]

⚠️ Search for existing issues first ⚠️

• I have searched the existing issues, and there is no existing issue for my problem

GPT-3 or GPT-4

• I am using Auto-GPT with GPT-3 (GPT-3.5)

Steps to reproduce 🕹

After issuing taks to agentes we get error message

and

Current behavior 😯

Python scripts breaks down

Expected behavior 🤔

Task query is processed

Your prompt 📝

# Paste your prompt here

[ruflair]

Hi!
Thank you for bringing this to my attention!
The issue lies with the lmql package (which is still very experimental). 0.0.6 and upwards breaks our code. I will push an update to the requirements.txt to reflect that we need 0.0.5.1 for now.
You can downgrade to 0.0.5.1 by running the command
pip install lmql==0.0.5.1
Let me know if this solves your problem!

[zepef]

Hello,

Thanks for the workaround it works great. We get 2 other problems now:

1) WARNING
To disable this warning, you can either: - Avoid using tokenizers before the fork if possible - Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)

We will try the workaround and I will keep tou updated.

2) ERROR (causing infinite loop)

Warning: Failed to parse AI output, attempting to fix.
If you see this warning frequently, it's likely that your prompt is confusing the AI. Try changing it up slightly.
Failed to fix AI output, telling the AI.
Error: Invalid JSON

On my PC/Win station everything works great. These 2 Mac configs are my client's, they do live in France I live in Madagascar. If you need me do not hesitate to get in touch (zepef@hotmail.com). I have 25+ years programming experience, I know Python/NLP/ML/LLM/Causal Inference/LangChain/VectorDatabases/Neo4j. I have been using Open AI products for the last 3 years. I wrote my first Expert System in Basic in 1989! I really like Multi-GPT.

Thanks for your outstanding work.

Pierre-Emmanuel